Merge branch 'bpf: expose bpf_{g,s}et_retval to more cgroup hooks'

Stanislav Fomichev says:

====================

Apparently, only a small subset of cgroup hooks actually falls
back to cgroup_base_func_proto. This leads to unexpected result
where not all cgroup helpers have bpf_{g,s}et_retval.

It's getting harder and harder to manage which helpers are exported
to which hooks. We now have the following call chains:

- cg_skb_func_proto
  - sk_filter_func_proto
    - bpf_sk_base_func_proto
      - bpf_base_func_proto

So by looking at cg_skb_func_proto it's pretty hard to understand
what's going on.

For cgroup helpers, I'm proposing we do the following instead:

  func_proto = cgroup_common_func_proto();
  if (func_proto) return func_proto;

  /* optional, if hook has 'current' */
  func_proto = cgroup_current_func_proto();
  if (func_proto) return func_proto;

  ...

  switch (func_id) {
  /* hook specific helpers */
  case BPF_FUNC_hook_specific_helper:
    return &xyz;
  default:
    /* always fall back to plain bpf_base_func_proto */
    bpf_base_func_proto(func_id);
  }

If this turns out more workable, we can follow up with converting
the rest to the same pattern.

v5:
- remove net/cls_cgroup.h include from patch 1/5 (Martin)
- move endif changes from patch 1/5 to 3/5 (Martin)
- don't define __weak protos, the ones in core.c suffice (Martin)

v4:
- don't touch existing helper.c helpers (Martin)
- drop unneeded CONFIG_CGROUP_BPF in bpf_lsm_func_proto (Martin)

v3:
- expose strtol/strtoul everywhere (Martin)
- move helpers declarations from bpf.h to bpf-cgroup.h (Martin)
- revise bpf_{g,s}et_retval documentation (Martin)
- don't expose bpf_{g,s}et_retval to cg_skb hooks (Martin)

v2:
- move everything into kernel/bpf/cgroup.c instead (Martin)
- use cgroup_common_func_proto in lsm (Martin)
====================

Signed-off-by: Alexei Starovoitov <ast@kernel.org>

3 files changed, 35 insertions, 5 deletions

diff --git a/include/linux/bpf-cgroup.h b/include/linux/bpf-cgroup.h
index 2bd1b5f8de9b..57e9e109257e 100644
--- a/include/linux/bpf-cgroup.h
+++ b/include/linux/bpf-cgroup.h
@@ -414,6 +414,11 @@ int cgroup_bpf_prog_detach(const union bpf_attr *attr,
 int cgroup_bpf_link_attach(const union bpf_attr *attr, struct bpf_prog *prog);
 int cgroup_bpf_prog_query(const union bpf_attr *attr,
 			  union bpf_attr __user *uattr);
+
+const struct bpf_func_proto *
+cgroup_common_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog);
+const struct bpf_func_proto *
+cgroup_current_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog);
 #else
 
 static inline int cgroup_bpf_inherit(struct cgroup *cgrp) { return 0; }
@@ -444,6 +449,18 @@ static inline int cgroup_bpf_prog_query(const union bpf_attr *attr,
 	return -EINVAL;
 }
 
+static inline const struct bpf_func_proto *
+cgroup_common_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
+{
+	return NULL;
+}
+
+static inline const struct bpf_func_proto *
+cgroup_current_func_proto(enum bpf_func_id func_id, const struct bpf_prog *prog)
+{
+	return NULL;
+}
+
 static inline int bpf_cgroup_storage_assign(struct bpf_prog_aux *aux,
 					    struct bpf_map *map) { return 0; }
 static inline struct bpf_cgroup_storage *bpf_cgroup_storage_alloc(
diff --git a/include/linux/bpf.h b/include/linux/bpf.h
index 39bd36359c1e..99fc7a64564f 100644
--- a/include/linux/bpf.h
+++ b/include/linux/bpf.h
@@ -2375,6 +2375,7 @@ extern const struct bpf_func_proto bpf_sock_map_update_proto;
 extern const struct bpf_func_proto bpf_sock_hash_update_proto;
 extern const struct bpf_func_proto bpf_get_current_cgroup_id_proto;
 extern const struct bpf_func_proto bpf_get_current_ancestor_cgroup_id_proto;
+extern const struct bpf_func_proto bpf_get_cgroup_classid_curr_proto;
 extern const struct bpf_func_proto bpf_msg_redirect_hash_proto;
 extern const struct bpf_func_proto bpf_msg_redirect_map_proto;
 extern const struct bpf_func_proto bpf_sk_redirect_hash_proto;
diff --git a/include/uapi/linux/bpf.h b/include/uapi/linux/bpf.h
index 7f87012b012e..644600dbb114 100644
--- a/include/uapi/linux/bpf.h
+++ b/include/uapi/linux/bpf.h
@@ -5085,17 +5085,29 @@ union bpf_attr {
  *
  * int bpf_get_retval(void)
  *	Description
- *		Get the syscall's return value that will be returned to userspace.
+ *		Get the BPF program's return value that will be returned to the upper layers.
  *
- *		This helper is currently supported by cgroup programs only.
+ *		This helper is currently supported by cgroup programs and only by the hooks
+ *		where BPF program's return value is returned to the userspace via errno.
  *	Return
- *		The syscall's return value.
+ *		The BPF program's return value.
  *
  * int bpf_set_retval(int retval)
  *	Description
- *		Set the syscall's return value that will be returned to userspace.
+ *		Set the BPF program's return value that will be returned to the upper layers.
+ *
+ *		This helper is currently supported by cgroup programs and only by the hooks
+ *		where BPF program's return value is returned to the userspace via errno.
+ *
+ *		Note that there is the following corner case where the program exports an error
+ *		via bpf_set_retval but signals success via 'return 1':
+ *
+ *			bpf_set_retval(-EPERM);
+ *			return 1;
+ *
+ *		In this case, the BPF program's return value will use helper's -EPERM. This
+ *		still holds true for cgroup/bind{4,6} which supports extra 'return 3' success case.
  *
- *		This helper is currently supported by cgroup programs only.
  *	Return
  *		0 on success, or a negative error in case of failure.
  *